- Email: [email protected]
Corrective Osteotomy for Malunited Diaphyseal Forearm Fractures Using Preoperative 3-Dimensional Planning and Patient-Specific Surgical Guides and Implants
SCIENTIFIC ARTICLE
Corrective Osteotomy for Malunited Diaphyseal Forearm Fractures Using Preoperative 3-Dimensional Planning and Patient-Specific Surgical Guides and Implants Ann-Maria Byrne, MD,* Bianca Impelmans, MD,* Veronique Bertrand, MS,† Annemieke Van Haver, PhD,‡ Frederik Verstreken, MD* Purpose Three-dimensional planning based on computed tomography images of the malunited and the mirrored contralateral forearm allows preoperative simulations of corrective osteotomies, the fabrication of patient-specific osteotomy guides, and custom-made 3-dimensional printed titanium plates. This study aims to assess the precision and clinical outcome of this technique. Methods This was a prospective pilot study with 5 consecutive patients. The mean age at initial injury was 11 years (range, 4e16 years), and the mean interval from the time of injury to the time of corrective surgery was 32 months (range, 7e107 months). Patient-specific osteotomy guides and custom-made plates were used for multiplanar corrective osteotomies of both forearm bones at the distal level in 1 patient and at the middle-third level in 4 patients. Patients were assessed before and after surgery after a mean follow-up of 42 months (range, 29e51 months). Results The mean planned angular corrections of the ulna and radius before surgery were 9.9 and 10.0 , respectively. The mean postoperative corrections obtained were 10.1 and 10.8 with corresponding mean errors in correction of 1.8 (range, 0.3 e5.2 ) for the ulna and 1.4 (range, 0.2 e3.3 ) for the radius. Forearm supination improved significantly from 47 (range, 25 e75 ) before surgery to 89 (range, 85 e90 ) at final review. Forearm pronation improved from 68 (range, 45 e84 ) to 87 (range, 82 e90 ). In addition, there was a statistically significant improvement in pain and grip strength. Conclusions This study demonstrates that 3-dimensional planned patient-specific guides and implants allow the surgeon to perform precise corrective osteotomies of complex multiplanar forearm deformities with satisfactory preliminary results. (J Hand Surg Am. 2017;-(-):1.e1-e12. Copyright Ó 2017 by the American Society for Surgery of the Hand. All rights reserved.) Type of study/level of evidence Therapeutic V. Key words Forearm, malunion, osteotomy, 3-dimensional planning, patient-specific guides and plates.
From the *Department of Orthopedic Surgery, Monica Hospital; the †More Foundation; and the ‡Monica Orthopedic Research Institute (More Institute), Antwerp, Belgium.
Corresponding author: Frederik Verstreken, MD, Department of Orthopedic Surgery, Monica Hospital, Florent Pauwelslei 1, 2100 Antwerp, Belgium; e-mail: [email protected].
Received for publication May 13, 2016; accepted in revised form June 1, 2017.
0363-5023/17/---0001$36.00/0 http://dx.doi.org/10.1016/j.jhsa.2017.06.003
Funding for this study was provided by the More Foundation. F.V. receives royalties from Materialise. The rest of the authors declare that they have no relevant conflicts of interest.
Ó 2017 ASSH
r
Published by Elsevier, Inc. All rights reserved.
r
1.e1
1.e2
P
FOREARM CORRECTION BASED ON COMPUTER PLANNING
FOREARM FRACTURES are routinely managed by closed methods with success in a large proportion of cases. Owing to potential remodeling during skeletal growth, fracture healing in a less than anatomical position is still compatible with unimpaired function.1 However, forearm malunions with angular deformity, displacement, and malrotation may result in functional impairment, pain, instability, and poor cosmetic appearance.2 Malunion rates between 15% and 35% have been reported following pediatric forearm fractures.3,4 Malunion with restriction of forearm rotation to less than 50% to 60% of normal is associated with considerable functional impairment and surgical intervention is recommended.2,5 To avoid long-term problems, the timing of corrective surgery is important, in relation to both skeletal maturity and time since the initial fracture.6 Corrective osteotomies performed in children younger than 10 years of age have demonstrated greater improvements in range of motion than those in older children.2,7 Increased range of motion and fewer complications have been reported with forearm corrective osteotomies performed within 12 months of the initial injury.8 Reconstructive surgery for malunited forearm fractures can be technically demanding. Precise forearm reconstruction may require simultaneous multiplanar correction of angulation and axial alignment, with restoration of ulnar variance and radioulnar joint congruity and stability, to obtain a pain-free, stable, and functional radioulnar joint. Conventional preoperative planning with 2-dimensional plain radiographs or cross-sectional imaging may not provide adequate information to fully appreciate the complexity of the 3-dimensional deformity, particularly with regards to rotational malalignment.9e11 Recent advances in computer technology allow accurate 3-dimensional deformity evaluation based on computed tomography (CT) data, and rapid prototyping technology allows development of patient-specific guides and implants.6,11e13 To establish a reliable surgical treatment for malunited forearm fractures, a 3-dimensional computer simulation system was used to compare the CT images of the malunited and mirrored contralateral forearm. The system allowed the reproduction of preoperative simulations during the surgical procedure with patient-specific osteotomy guides and custom-made implants, where standard implants failed to match the surface contours of the osteotomized bones during the planning simulation. This study hypothesizes that preoperative 3-dimensional planning combined with patient-specific guides and
plates can correct angular deformities of the forearm bones to within 5 of the contralateral side.
EDIATRIC
J Hand Surg Am.
MATERIALS AND METHODS Patients From December 2011 to May 2013, 5 skeletally immature patients with diaphyseal forearm malunions (4 at the middle-third level, 1 at the distal level) presented as tertiary referrals and were enrolled in this prospective study (Table 1). The inclusion criteria were skeletally immature patients with decreased forearm rotation secondary to osseous deformity of the diaphysis of both forearm bones and incongruence between standard plate constructs and the anatomical contour of the corrected radius and ulna on the preoperative computer simulation. Exclusion criteria included medical comorbidities that could interfere with study participation, any involvement of adjacent joints, previous injury to the contralateral forearm, and cases in which standard fixation plates could be used. Initially, 4 patients had been managed with closed reduction and cast immobilization; 1 patient was treated with closed reduction and intramedullary nailing. Prior to deciding on surgical forearm correction, sequential orthogonal radiographs were performed to confirm the presence of osseous union and that no further remodeling had occurred (Fig. 1). Corrective surgery was indicated for loss of forearm rotation in 1 patient, and loss of forearm rotation with wrist or forearm pain in 4 patients. All patients underwent corrective osteotomy of both the radius and the ulna using preoperative 3-dimensional planning and custom-made patient-specific osteotomy guides and plate constructs (Fig. 2; Video A; available on the Journal’s Web site at www.jhandsurg.org). All patients and their parents gave informed consent before surgery and institutional review board approval was provided for this study. The study included 3 male and 2 female patients with a mean age of 11 years (range, 4e16 years) at the time of initial injury (Table 1). At the time of corrective surgery, the mean age was 13 years (range, 7e17 years). The median interval from the time of injury to the time of corrective surgery was 16 months (range, 7e107 months). None of the patients were lost to follow-up or excluded from the study after initial enrollment. Preoperative planning The affected and contralateral forearms had CT imaging (Somatom Sensation 64 CT scanner: slice thickness, 0.6 mm; slice increment, 0.6 mm; image r
Vol. -, - 2017
38 Radius opening wedge Ulna opening wedge Distal M
15.8
17.1
Wiring
Design and manufacturing of the custom-made guides and plate constructs The virtual model simulations allowed planning of optimal plate construct position on the corrected radius and ulna according to the chosen surgical approach and required osteotomy. Because standard plate constructs failed to fit the contour of the corrected bone in these cases, custom-made plates were designed for the radius and ulna (Mobelife, Leuven, Belgium). A smooth peg at both ends of the plate was incorporated into the design of the plate to facilitate correct positioning of the plate and bone fragments (Fig. 3). To re-create the preoperative simulation during surgery, custom-made guides with guided drill holes and osteotomy cutting slots were designed using commercially available software (3-Matic). These virtual guides were then generated by 3-dimensional printing technology as medical grade polyamide models and sterilized prior to surgery. The plates were 3-dimensionally printed in titanium grade 23 (Ti6Al4V ELI; (LayerWise N.V., Leuven, Belgium) by a selective laser melting installation and were
5 J Hand Surg Am.
1.e3
matrix, 512 512 pixels; pixel size, 0.269 mm; Siemens Healthcare, Erlangen, Germany) from the elbow to the carpometacarpal joints. The CT data were saved in a standard format (DICOM [Digital Imaging and Communications in Medicine]). The formatted files were then imported into the computer simulation software (Mimics and SurgiCase; Materialise N.V., Leuven, Belgium) to segment the CT images and generate 3-dimensional virtual models of both forearm bones of the malunited and mirrored contralateral forearm. These virtual models allow multiplanar visualization and processing with 3-dimensional annotation and measurement. The degree of deformity was evaluated with 3-Matic software (Materialise N.V.) by using the angles between the principal axes of inertia (PAI) of the proximal and distal radius and ulna, adapting techniques previously used and validated by Victor and colleagues14 (Fig. 3). The rotational and translational displacements were calculated as the difference between the affected bone model and the mirror model of the normal bone superimposed proximally to distally. Based on these 3-dimensional models, virtual osteotomies were simulated to align the malunited forearm with the contralateral side, thus producing a precise correction of angular and rotational malalignment and restoring ulnar variance (Video A; available on the Journal’s Web site at www.jhandsurg.org).
15
51 Radius opening wedge Ulna closing wedge M 4
6.6
7.2
Conservative
Midthird
7
41 Radius oblique opening wedge Ulna opening wedge M 3
13.1
14.2
Conservative
Midthird
14
29 Radius closing wedge Ulna oblique F 2
12.8
14.1
Conservative
Midthird
16
51 107 Radius opening wedge Ulna opening wedge Midthird Conservative 13.4 4.4 F 1
Age at Time Fracture (y) Sex Case
TABLE 1.
Patient Characteristics
Age at Time Osteotomy (y)
Initial Treatment
Level of Malunion
Osteotomy
Interval To Osteotomy (mo)
Follow-Up (mo)
FOREARM CORRECTION BASED ON COMPUTER PLANNING
r
Vol. -, - 2017
1.e4
FOREARM CORRECTION BASED ON COMPUTER PLANNING
FIGURE 1: Preoperative A and postoperative B radiographs of a patient with a malunited fracture of both forearm bones.
and guides and V1,500 (U.S. $1,600) for the 2 implants.
FIGURE 2: Patient-specific constructs.
osteotomy
guides
and
Surgical technique Based on the preoperative 3-dimensional virtual planning, all patients underwent an opening, closing, or oblique osteotomy of the radius and ulna to obtain the best possible correction (Table 1). The patient was placed in the supine position with the affected arm on an arm table with a tourniquet applied. In all cases, the ulna was corrected first with a medial approach between the flexor carpi ulnaris and the extensor carpi ulnaris. The polyamide guide was precisely positioned on the malunited ulna using the surface anatomy of the bone (Video A; available on the Journal’s Web site at www.jhandsurg.org). A polyamide 3-dimensional printed copy of the ulna with attached guide and fluoroscopy are used to confirm the correct position of the guide (Fig. 4). The first drill guide was fixed to the ulna with 1.3-mm K-wires and the ulna was drilled through the guiding drill holes. Then the osteotomy guide was placed in position with the use of the same K-wires and the osteotomy cuts were made with an oscillating saw (TPS system; Stryker Orthopaedics, Mahwah, NJ). The guide and K-wires are removed and the patient-specific plate was positioned by inserting the smooth pegs in the predrilled holes, providing an initial reduction and fixation.
plate
designed according to dynamic compression plate principles. Total costs per patient currently amount to approximately V2,500 (U.S. $2,700) for planning J Hand Surg Am.
r
Vol. -, - 2017
FOREARM CORRECTION BASED ON COMPUTER PLANNING
1.e5
FIGURE 3: Study design. Preoperative CT scans are loaded in Mimics A and are applied to perform the 3-dimensioonal planning B for the corrective osteotomy with patient-specific guides and plates. After surgery, a postoperative x-ray is performed C, and the proximal parts of the bone models of the preoperative A, the 3-dimensional planning B, and the postoperative C status are superimposed for the radiographic analysis in the plane of the lateral and AP views of the postoperative radiograph in 3-Matic.
Subsequently, the plate was further fixed with 4 titanium nonlocking, fully threaded cortical screws (2.7 mm diameter in 2 patients, 3.5 mm diameter in 3 patients), providing fixation in 6 cortices both proximally and distally of the osteotomy (Synthes, West Chester, PA). The smooth pegs of the patientspecific plates were 0.5-mm undersized compared with the drill holes in the bone, allowing additional compression by the screws. The correction was verified under fluoroscopy. For the radius, a volar Henry approach was used and a similar procedure was performed. Additional soft tissue release was not required for correction of the deformity.
home exercise program under parent supervision. Full load-bearing and contact sports was allowed once osseous union was well established. Because of the age of the patients, plate removal was planned routinely in all patients. Radiographic evaluation Radiographic evaluation of the angular correction was performed using the preoperative CT-based 3-dimensional models, the planned corrected 3dimensional models, and the postoperative orthogonal radiographs of the osteotomized forearm (Fig. 1B). After surgery, CT scans were not performed to avoid unnecessary radiation of these skeletally immature patients. To allow radiological comparison of the 3 phases (preoperative, planning, postoperative), the 3-dimensional models of the
Postoperative management The forearm was protected with an above-elbow orthosis for 2 weeks, followed by rehabilitation in a J Hand Surg Am.
r
Vol. -, - 2017
1.e6
FOREARM CORRECTION BASED ON COMPUTER PLANNING
procedure a single time for the preoperative, planned, and postoperative phases. In the Results section, the planned angulations (based on the contralateral side) are reported as positive values. The pre- and postoperative angulations are reported positive when the distal bone was deviating in the same direction as the planned angulation. Overcorrection errors are reported as positive and undercorrection errors are reported as negative. In addition, postoperative ulnar variance was assessed by comparing the ulnar variance of the affected side with the contralateral side on the postoperative radiographs. Clinical evaluation An experienced hand therapist (V.B.) performed independent clinical assessments before surgery, at 1 year, and at a minimum of 2 years follow-up of range of motion, grip strength, and pain. Mean follow-up was 42 months (range, 29e51 months). The ranges of wrist flexion, wrist extension, forearm pronation, and forearm supination were measured in degrees with a goniometer. Grip strength was measured in Newtons (N) with a JAMAR isometric handgrip dynamometer (Sammons Preston Rolyan, Bolingbrook, IL). The JAMAR handle was set to position 2 for all subjects. Grip strength was evaluated alternately 3 times for the right hand and 3 times for the left hand with the elbow flexed to 90 and the forearm and wrist in neutral position. The average score was used and a 15% correction for limb dominance was performed for right-handed individuals. Pain scores were measured between 0 (no pain) and 10 (maximum pain) with a visual analog scale.
FIGURE 4: Confirmation of the correct position of the guide by making a fluoroscopy B of the patient’s ulna, together with a polyamide 3-dimensional printed copy of the ulna, attached to the guide A.
preoperative malunion and planned correction were evaluated in the same views as the postoperative radiographs. The DICOM files of the postoperative radiographs and the 3-dimensional models of the preoperative and corrected radius and ulna were imported to the Mimics and 3-Matic software. The 3-dimensional models were set on transparent mode and were individually positioned such that the proximal contours of the 3-dimensional bone models fitted the contours of the radius and ulna on the radiograph (Fig. 3). This alignment procedure was performed once by 3 independent observers (all 3 were orthopedic surgeons), blinded to the measurements of the other evaluators. After aligning the 3-dimensional bone models on the radiographs, the PAI of the proximal and distal radius and ulna were projected on the plane of the radiograph, allowing a quantification of the preoperative, planned, and postoperative angulation of the ulna and radius in the anteroposterior (AP) and lateral planes (Fig. 3). The angulations were then quantified as the angles between the projected PAI of the proximal and the distal part of the radius and ulna. The 3 observers performed this measurement J Hand Surg Am.
Statistical methods Owing to the small sample size, nonparametric statistical methods were employed and the differences between the preoperative and the postoperative range of motion, grip strength, and visual analog scale pain scores were determined using the related samples Wilcoxon signed-rank test. The intraclass correlation coefficient was used to estimate the interobserver reliability of the radiological measurements among 3 independent observers. A P value of .05 or less was considered statistically significant. Ninety-five percent confidence intervals are indicated in the graphical presentation of the clinical and radiographic measurements. Results were expressed as mean (range). Bar graphs were used to indicate means with 95% confidence intervals. r
Vol. -, - 2017
1.e7
FOREARM CORRECTION BASED ON COMPUTER PLANNING
Ulna
Radius
Average Angulation (°) With 95% CI
30
Preoperative Planned Postoperative Correction error
20
10
0
–10
–20
AP
Lateral
AP
Lateral
FIGURE 5: Radiographic measurements. Average angulation of the preoperative, planned, and postoperative ulna and radius, and error in angular correction, measured in the plane of the AP and lateral radiographs with 95% confidence interval (95% CI) bars. Planned angulation was defined positive for all subjects, overcorrection errors are reported as positive, and undercorrection errors as negative values.
RESULTS Radiographic results For the ulna, the preoperative angulation varied between e19.7 and 23.3 in the lateral view and between e13.4 and 20.8 in the AP view. After correction, these deviating angulations were normalized to a range between 0.9 and 6.6 in the lateral view and between 0.6 and 5.2 in the AP view corresponding to the natural curvature of the contralateral bone (Fig. 5). The difference between the planned and the actual correction of the ulna was on average 1.6 (range, 0.6 e4.0 ) in the lateral view and 2.1 (range, 0.3 e5.2 ) in the AP view (Fig. 5; Table S1; available on the Journal’s Web site at www.jhandsurg.org). For the radius, the preoperative angulation varied between e12.5 and 18.1 in the lateral view and between e14.1 and 19.9 in the AP view (Fig. 5). After correction, these deviating angulations were normalized to a range between 1.7 and 9.7 in the lateral view and between 3.1 and 8.4 in the AP view (Fig. 5). The difference between the planned and the actual correction of the radius was on average 1.0 (range, 0.2 e2.3 ) in the lateral view and 1.8 (range, 0.2 e3.3 ) in the AP view (Fig. 5; Table S1; available on the Journal’s Web site at www.jhandsurg.org).
J Hand Surg Am.
The interobserver reliability of the radiological measurements of the alignment of the preoperative, planned, and postoperative models of the ulna and radius was 0.99 (the intraclass correlation coefficient, calculated for 180 observations [3 observers 5 patients 2 radiographic planes 2 bones 3 phases]). For all patients, the postoperative difference in ulnar variance between the affected and the contralateral side was within 1 mm. Clinical results One patient was found to have insufficient osseous union at the time of ulnar plate removal and had immediate replating with a compression plate, followed by successful union. The other 4 patients healed well without complications. All patients showed restricted forearm rotation before surgery and regained close to normal forearm rotation following the osteotomy without release of the interosseous membrane (Fig. 6). The clinical assessments following the corrective surgery showed important improvements in pain score (P < .05), grip strength (P < .05), flexion (P ¼ .07), forearm supination (P < .05), and forearm pronation (P ¼ .08) (Fig. 6).
r
Vol. -, - 2017
1.e8
FOREARM CORRECTION BASED ON COMPUTER PLANNING
FIGURE 6: Clinical measurements. A Pre- and postoperative average scores for pain, range of motion, and grip strength with 95% confidence interval (95% CI) bars. B Pre- and postoperative average scores for range of motion and grip strength compared with the uninjured contralateral forearm with 95% CI bars. A value of 100% means no difference compared with the contralateral side.
DISCUSSION This study investigated the accuracy of the angular correction and clinical outcome of preoperative CTbased 3-dimensional planning and patient-specific guides and plates for the treatment of pediatric forearm malunions. Earlier clinical and cadaver studies have demonstrated that forearm malunions may cause severe restriction of forearm rotation.15e18 Angular deformities in the middle or distal third of the forearm16 and combined multidirectional angular deformities of the radius and ulna17 have been shown to have a J Hand Surg Am.
higher impact on the range of motion than proximal and unidirectional deformities. Multiplanar correction of forearm malunions with restoration of ulnar variance requires meticulous preoperative planning. Translating these plans into an accurate surgical correction can be challenging, and a failure to appreciate the magnitude of the required correction, resulting in inaccurate surgical osteotomies, may lead to inferior clinical results, particularly in complex deformities of the forearm. Previous studies have used 2-dimensional orthogonal radiographs in the preoperative planning.
r
Vol. -, - 2017
FOREARM CORRECTION BASED ON COMPUTER PLANNING
However, assessment of 3-dimensional deformities with 2-dimensional images has significant limitations.2,6,7,9,10,12,18 Owing to difficulty in estimating axial malunions with 2-dimensional radiographs, it has previously been recommended to assess axial malunions before surgery with magnetic resonance imaging and fluoroscopy coupled with goniometry to plan corrective forearm osteotomies.19,20 Nagy et al19 performed orthogonal radiography and axial magnetic resonance imaging as preoperative planning for 17 forearm osteotomy patients. They noted that residual angulation or axial malunion after forearm osteotomies resulted from incorrect planning or inaccuracy in the surgical correction. Whereas the postoperative mean arc of forearm motion improved from 82 to 124 , the overall improvement in motion for patients with predominantly a pronation deficit was poor, and the mean residual restricted forearm motion was 44 . They concluded that combining correction of angular and axial deformities in the forearm was difficult because rotation around an oblique osteotomy induces a change in angulation. More recently, studies have described the use of 3dimensional technology to improve preoperative planning and develop customized cutting guides for osteotomies of the femur, tibia, humerus, and forearm bones.10,11,13,14,21 Murase et al10 investigated 10 patients with malunited diaphyseal forearm fractures who were managed by corrective osteotomy with a custom-made guide based on computer simulations. The average angle of deformity improved from 16 (range, 5 e33 ) before surgery to 1 (range, 0 e3 ) after corrective osteotomy. The average range of forearm pronation and supination significantly improved from 60 and 19 , respectively, to 82 and 73 . One patient, who had corrective surgery 9 years after initial injury, had persistent restricted forearm supination despite good radiological correction. Miyake and colleagues11 reported on 20 forearm malunion patients using the same technique. The average angle of deformity improved from 21 (range, 12 e35 ) before surgery to 1 (range, 0 e4 ) after surgery with an improvement in the mean arc of forearm motion from 76 to 152 . However, forearm supination was still restricted by 70 in 3 patients who had been younger than 10 years at the time of their initial injury and who had long-standing malunion for 96 months or longer. It was surmised that changes in joint configurations and soft tissue contractures during the long delay prior to correction caused residual restrictions in forearm supination. In the current study, we evaluated 5 carefully selected pediatric patients. All patients were skeletally J Hand Surg Am.
1.e9
immature, showed complex malunions of the radius and ulna, and showed incongruence between the standard plates and the bones. Therefore, these patients did not require only customized cutting guides but also patient-specific plates to enhance precise reconstruction of the radius and ulna. For the preoperative planning, we identified the most favorable pattern of corrective osteotomy by superimposing the 3-dimensional models of the malunited and contralateral forearms. Following the planning stage, the patient-specific guides and plates enabled precise intraoperative reconstruction of the preoperative virtual osteotomy simulation, with no intraoperative issues. Compared with the earlier findings of Murase et al10 and Miyake et al,11 the accuracy obtained in this study is very similar. The clinical results of our 5 patients were very good; none of the patients showed restricted forearm range of motion after follow-up, whereas Miyake and colleagues11 still found restricted forearm supination in 3 young patients with long-standing malunions of 96 months or longer. This study also included 1 patient with a longstanding malunion of 107 months who was 4 years old at the time of the injury and 10 at the time of corrective surgery. In this patient, potential changes in joint configurations and soft tissue contractures did not cause any residual restriction of forearm supination. The other 4 patients, aged between 6 and 16 years at the time of injury, had a corrective osteotomy with a delay between 7 and 16 months. The short duration between the malunion and the corrective osteotomy of these patients, as well as their young age, may have contributed to the good outcome of the intervention. To elaborate on the relation between surgical delay and clinical outcome, continued research on larger study groups, including adult patients with long-standing malunions, is necessary. Whereas Murase et al10 and Miyake et al11 used standard plates, we additionally used patient-specific plates. Several studies have demonstrated that the mechanical properties of selective laser melting 3dimensional printed titanium are comparable with, or better than, the standard Ti6Al4V material,22,23 but its use as patient-specific bone plates has only been described for a limited number of surgical applications: for hallux valgus deformities,24 complex pelvic fractures,25 and craniofacial applications.26 The 3-dimensional printed patient-specific forearm plates in this study permitted a perfect fit on the corrected bones and provided the possibility of adding pegs to the plates. These features facilitated the reduction of the osteotomy and allowed precise r
Vol. -, - 2017
1.e10
FOREARM CORRECTION BASED ON COMPUTER PLANNING
fixation of the bony fragments in the planned position. Intraoperative bending of standard plates is an alternative option,21 but is time consuming and the result depends on the complexity of the anatomy and the surgeon’s skill. Whether the clinical benefits of patient-specific implants outweigh the additional cost remains to be determined. The authors acknowledge several weaknesses of this study. In this preliminary series, we assessed the accuracy of corrective osteotomies using 3dimensional technology without comparing the results with a control group without 3-dimensional technology. During the study period, only 5 patients met the inclusion criteria. Because of the small sample size, nonparametric statistical tests were performed that inevitably decreased the power of the study. As a result, this study may fail to detect statistical differences between the pre- and the postoperative situation. Nevertheless, this preliminary study demonstrated statistically significant improvements for several parameters. The contralateral forearm was used to calculate the magnitude of 3-dimensional correction, although forearm-toforearm variability in the normal population is well recognized.20 Future computerized simulations using mathematical models might be developed to avoid imaging the opposite forearm as well as incorporating the biomechanical effects of the interosseous membrane20 but, at present, this is not available. There was a potential bias toward patients with complex malunions who were motivated to try new technology and participate in this study to assess the potential advantage of the 3-dimensional printed patientspecific guides and plates. Interosseous membrane release was not performed; thus, outcomes were entirely dependent on correction of the osseous deformities. To avoid unnecessary radiation, postoperative CT scans were not performed. Instead, orthogonal radiographs were performed that allowed postoperative quantification of the angular correction but not of the torsional correction. Although the results of this small series are encouraging, continued research on larger study groups is required to further evaluate the potential of this technology. We believe this technique represents an exciting and promising new treatment option in the management of complex forearm malunions.
Lucas H.B. Walschot, MD, PhD, who contributed to the manuscript revision. Funding was received from the More Foundation. REFERENCES 1. Hughston JC. Fracture of the forearm in children. J Bone Joint Surg Am. 1962;44(8):1678e1693. 2. van Geenen RC, Besselaar PP. Outcome after corrective osteotomy for malunited fractures of the forearm sustained in childhood. J Bone Joint Surg Br. 2007;89(2):236e239. 3. Fuller DJ, McCullough CJ. Malunited fractures of the forearm in children. J Bone Joint Surg Br. 1982;64(3):364e367. 4. Schmittenbecher PP. State-of-the-art treatment of forearm shaft fractures. Injury. 2005;36(Suppl 1):A25eA34. 5. Schemitsch EH, Richards RR. The effect of malunion on functional outcome after plate fixation of fractures of both bones of the forearm in adults. J Bone Joint Surg Am. 1992;74(7):1068e1078. 6. Jayakumar P, Jupiter JB. Reconstruction of malunited diaphyseal fractures of the forearm. Hand (N Y). 2014;9(3):265e273. 7. Price CT, Knapp DR. Osteotomy for malunited forearm shaft fractures in children. J Pediatr Orthop. 2006;26(2):193e196. 8. Trousdale RT, Linscheid RL. Operative treatment of malunited fractures of the forearm. J Bone Joint Surg Am. 1995;77(6):894e902. 9. Bilic R, Zdravkovic V, Boljevic Z. Osteotomy for deformity of the radius. Computer-assisted three-dimensional modelling. J Bone Joint Surg Br. 1994;76(1):150e154. 10. Murase T, Oka K, Moritomo H, Goto A, Yoshikawa H, Sugamoto K. Three-dimensional corrective osteotomy of malunited fractures of the upper extremity with use of a computer simulation system. J Bone Joint Surg Am. 2008;90(11):2375e2389. 11. Miyake J, Murase T, Oka K, Moritomo H, Sugamoto K, Yoshikawa H. Computer-assisted corrective osteotomy for malunited diaphyseal forearm fractures. J Bone Joint Surg Am. 2012;94(20): e150. 12. Jupiter JB, Ruder J, Roth DA. Computer-generated bone models in the planning of osteotomy of multidirectional distal radius malunions. J Hand Surg Am. 1992;17(3):406e415. 13. Oka K, Murase T, Moritomo H, Goto A, Sugamoto K, Yoshikawa H. Corrective osteotomy using customized hydroxyapatite implants prepared by preoperative computer simulation. Int J Med Robot. 2010;6(2):186e193. 14. Victor J, Premanathan A. Virtual 3D planning and patient specific surgical guides for osteotomies around the knee: a feasibility and proof-of-concept study. Bone Joint J. 2013;95-B(11 Suppl A): 153e158. 15. Morrey BF, Askew LJ, An KN, et al. A biomechanical study of normal functional elbow motion. J Bone Joint Surg Am. 1981;63(6): 872e877. 16. Sarmiento A, Ebramzadeh E, Brys D, Tarr R. Angular deformities and forearm function. J Orthop Res. 1992;10(1):121e133. 17. Tarr RR, Garfinkel AI, Sarmiento A. The effects of angular and rotational deformities of both bones of the forearm. An in vitro study. J Bone Joint Surg Am. 1984;66(1):65e70. 18. Dumont CE, Thalmann R, Macy JC. The effect of rotational malunion of the radius and the ulna on supination and pronation. J Bone Joint Surg Br. 2002;84(7):1070e1074. 19. Nagy L, Jankauskas L, Dumont CE. Correction of forearm malunion guided by the preoperative complaint. Clin Orthop Relat Res. 2008;466(6):1419e1428. 20. Dumont CE, Pfirrmann CW, Ziegler D, Nagy L. Assessment of radial and ulnar torsion profiles with cross-sectional magnetic resonance imaging: a study of volunteers. J Bone Joint Surg Am. 2006;88(7): 1582e1588. 21. Kataoka T, Oka K, Miyake J, Omori S, Tanaka H, Murase T. 3Dimensional prebent plate fixation in corrective osteotomy of malunited upper extremity fractures using a real-sized plastic bone model
ACKNOWLEDGMENTS We are grateful to Kristien Vuylsteke (clinical research coordinator at More Foundation), who took care of patient follow-up and data collection and to
J Hand Surg Am.
r
Vol. -, - 2017
FOREARM CORRECTION BASED ON COMPUTER PLANNING
24. Smith KE, Dupont KM, Safranski DL, et al. Use of 3D printed bone plate in novel technique to surgically correct hallux valgus deformities. Tech Orthop. 2016;31(3):181e189. 25. Wang D, Wang Y, Wu S, et al. Customized a Ti6Al4V bone plate for complex pelvic fracture by selective laser melting. Materials. 2017;10(1):35. 26. Jacobs C, Lin A. A new classification of three-dimensional printing technologies: systematic review of three-dimensional printing for patient-specific craniomaxillofacial surgery. Plast Reconstr Surg. 2017;139(5):1211e1220.
prepared by preoperative computer simulation. J Hand Surg Am. 2013;38(5):909e919. 22. Murr LE, Quinones SA, Gaytana SM, et al. Microstructure and mechanical behavior of Ti-6Al-4V produced by rapid-layer manufacturing, for biomedical applications. J Mech Behav Biomed Mater. 2009;2(1):20e32. 23. Rafi HK, Karthik NV, Gong H, et al. Microstructures and mechanical properties of Ti6Al4V parts fabricated by selective laser melting and electron beam melting. J Mater Eng Perform. 2013;22(12): 3872e3883.
J Hand Surg Am.
1.e11
r
Vol. -, - 2017
1.e12
FOREARM CORRECTION BASED ON COMPUTER PLANNING
TABLE S1.
Clinical and Radiographic Measurements per Patient Patient
Clinical Measurements
1
2
3
4
5
Preoperative
80/55
82/75
45/25
84/31
50/50
Postoperative
90/90
82/90
85/90
90/90
88/85
Preoperative
135.0
157.0
70
115
100
Postoperative
180.0
172.0
175
180
173
Preoperative
32.0
18.3
29.3
10
30
Postoperative
36.0
25.0
36
14.7
30.3
Preoperative
3.5
3.4
1.2
3
5
Postoperative
1.0
0.0
0
0
2.6
e13.4
5.7
11.3
20.8
2.3
Forearm rotation: pronation/supination ( )
Arc of rotation ( )
Grip strength (kg)
Pain scores (VAS)
Radiographic Measurements Ulna—AP view ( ) Preoperative Planned
1.9
6.4
6.3
5.2
1.0
Postoperative
0.6
5.2
1.1
2.8
1.4
Error
1.3
1.3
5.2
2.4
0.3
Ulna—lateral view ( ) Preoperative
7.0
e0.1
e19.7
23.2
6.4
Planned
0.3
5.8
5.2
5.5
0.4
Actual
0.9
6.6
4.3
4.0
4.4
Error
0.6
0.8
0.9
1.5
4.0
Preoperative
16.2
e5.5
5.7
e14.1
19.9
Planned
9.5
4.9
8.0
0.4
4.6
Actual
8.4
7.0
4.7
3.1
4.4
Error
1.1
2.1
3.3
2.7
0.2
Preoperative
9.5
11.2
18.1
e12.5
3.0
Planned
4.2
5.4
2.4
10.0
4.3
Actual
4.7
3.1
1.7
9.8
5.6
Error
0.5
2.3
0.7
0.2
1.3
Radius—AP view ( )
Radius—lateral view ( )
VAS, visual analog scale.
J Hand Surg Am.
r
Vol. -, - 2017
Corrective Osteotomy for Malunited Diaphyseal Forearm Fractures Using Preoperative 3-Dimensional Planning and Patient-Specific Surgical Guides and Implants Ann-Maria Byrne, MD,* Bianca Impelmans, MD,* Veronique Bertrand, MS,† Annemieke Van Haver, PhD,‡ Frederik Verstreken, MD* Purpose Three-dimensional planning based on computed tomography images of the malunited and the mirrored contralateral forearm allows preoperative simulations of corrective osteotomies, the fabrication of patient-specific osteotomy guides, and custom-made 3-dimensional printed titanium plates. This study aims to assess the precision and clinical outcome of this technique. Methods This was a prospective pilot study with 5 consecutive patients. The mean age at initial injury was 11 years (range, 4e16 years), and the mean interval from the time of injury to the time of corrective surgery was 32 months (range, 7e107 months). Patient-specific osteotomy guides and custom-made plates were used for multiplanar corrective osteotomies of both forearm bones at the distal level in 1 patient and at the middle-third level in 4 patients. Patients were assessed before and after surgery after a mean follow-up of 42 months (range, 29e51 months). Results The mean planned angular corrections of the ulna and radius before surgery were 9.9 and 10.0 , respectively. The mean postoperative corrections obtained were 10.1 and 10.8 with corresponding mean errors in correction of 1.8 (range, 0.3 e5.2 ) for the ulna and 1.4 (range, 0.2 e3.3 ) for the radius. Forearm supination improved significantly from 47 (range, 25 e75 ) before surgery to 89 (range, 85 e90 ) at final review. Forearm pronation improved from 68 (range, 45 e84 ) to 87 (range, 82 e90 ). In addition, there was a statistically significant improvement in pain and grip strength. Conclusions This study demonstrates that 3-dimensional planned patient-specific guides and implants allow the surgeon to perform precise corrective osteotomies of complex multiplanar forearm deformities with satisfactory preliminary results. (J Hand Surg Am. 2017;-(-):1.e1-e12. Copyright Ó 2017 by the American Society for Surgery of the Hand. All rights reserved.) Type of study/level of evidence Therapeutic V. Key words Forearm, malunion, osteotomy, 3-dimensional planning, patient-specific guides and plates.
From the *Department of Orthopedic Surgery, Monica Hospital; the †More Foundation; and the ‡Monica Orthopedic Research Institute (More Institute), Antwerp, Belgium.
Corresponding author: Frederik Verstreken, MD, Department of Orthopedic Surgery, Monica Hospital, Florent Pauwelslei 1, 2100 Antwerp, Belgium; e-mail: [email protected].
Received for publication May 13, 2016; accepted in revised form June 1, 2017.
0363-5023/17/---0001$36.00/0 http://dx.doi.org/10.1016/j.jhsa.2017.06.003
Funding for this study was provided by the More Foundation. F.V. receives royalties from Materialise. The rest of the authors declare that they have no relevant conflicts of interest.
Ó 2017 ASSH
r
Published by Elsevier, Inc. All rights reserved.
r
1.e1
1.e2
P
FOREARM CORRECTION BASED ON COMPUTER PLANNING
FOREARM FRACTURES are routinely managed by closed methods with success in a large proportion of cases. Owing to potential remodeling during skeletal growth, fracture healing in a less than anatomical position is still compatible with unimpaired function.1 However, forearm malunions with angular deformity, displacement, and malrotation may result in functional impairment, pain, instability, and poor cosmetic appearance.2 Malunion rates between 15% and 35% have been reported following pediatric forearm fractures.3,4 Malunion with restriction of forearm rotation to less than 50% to 60% of normal is associated with considerable functional impairment and surgical intervention is recommended.2,5 To avoid long-term problems, the timing of corrective surgery is important, in relation to both skeletal maturity and time since the initial fracture.6 Corrective osteotomies performed in children younger than 10 years of age have demonstrated greater improvements in range of motion than those in older children.2,7 Increased range of motion and fewer complications have been reported with forearm corrective osteotomies performed within 12 months of the initial injury.8 Reconstructive surgery for malunited forearm fractures can be technically demanding. Precise forearm reconstruction may require simultaneous multiplanar correction of angulation and axial alignment, with restoration of ulnar variance and radioulnar joint congruity and stability, to obtain a pain-free, stable, and functional radioulnar joint. Conventional preoperative planning with 2-dimensional plain radiographs or cross-sectional imaging may not provide adequate information to fully appreciate the complexity of the 3-dimensional deformity, particularly with regards to rotational malalignment.9e11 Recent advances in computer technology allow accurate 3-dimensional deformity evaluation based on computed tomography (CT) data, and rapid prototyping technology allows development of patient-specific guides and implants.6,11e13 To establish a reliable surgical treatment for malunited forearm fractures, a 3-dimensional computer simulation system was used to compare the CT images of the malunited and mirrored contralateral forearm. The system allowed the reproduction of preoperative simulations during the surgical procedure with patient-specific osteotomy guides and custom-made implants, where standard implants failed to match the surface contours of the osteotomized bones during the planning simulation. This study hypothesizes that preoperative 3-dimensional planning combined with patient-specific guides and
plates can correct angular deformities of the forearm bones to within 5 of the contralateral side.
EDIATRIC
J Hand Surg Am.
MATERIALS AND METHODS Patients From December 2011 to May 2013, 5 skeletally immature patients with diaphyseal forearm malunions (4 at the middle-third level, 1 at the distal level) presented as tertiary referrals and were enrolled in this prospective study (Table 1). The inclusion criteria were skeletally immature patients with decreased forearm rotation secondary to osseous deformity of the diaphysis of both forearm bones and incongruence between standard plate constructs and the anatomical contour of the corrected radius and ulna on the preoperative computer simulation. Exclusion criteria included medical comorbidities that could interfere with study participation, any involvement of adjacent joints, previous injury to the contralateral forearm, and cases in which standard fixation plates could be used. Initially, 4 patients had been managed with closed reduction and cast immobilization; 1 patient was treated with closed reduction and intramedullary nailing. Prior to deciding on surgical forearm correction, sequential orthogonal radiographs were performed to confirm the presence of osseous union and that no further remodeling had occurred (Fig. 1). Corrective surgery was indicated for loss of forearm rotation in 1 patient, and loss of forearm rotation with wrist or forearm pain in 4 patients. All patients underwent corrective osteotomy of both the radius and the ulna using preoperative 3-dimensional planning and custom-made patient-specific osteotomy guides and plate constructs (Fig. 2; Video A; available on the Journal’s Web site at www.jhandsurg.org). All patients and their parents gave informed consent before surgery and institutional review board approval was provided for this study. The study included 3 male and 2 female patients with a mean age of 11 years (range, 4e16 years) at the time of initial injury (Table 1). At the time of corrective surgery, the mean age was 13 years (range, 7e17 years). The median interval from the time of injury to the time of corrective surgery was 16 months (range, 7e107 months). None of the patients were lost to follow-up or excluded from the study after initial enrollment. Preoperative planning The affected and contralateral forearms had CT imaging (Somatom Sensation 64 CT scanner: slice thickness, 0.6 mm; slice increment, 0.6 mm; image r
Vol. -, - 2017
38 Radius opening wedge Ulna opening wedge Distal M
15.8
17.1
Wiring
Design and manufacturing of the custom-made guides and plate constructs The virtual model simulations allowed planning of optimal plate construct position on the corrected radius and ulna according to the chosen surgical approach and required osteotomy. Because standard plate constructs failed to fit the contour of the corrected bone in these cases, custom-made plates were designed for the radius and ulna (Mobelife, Leuven, Belgium). A smooth peg at both ends of the plate was incorporated into the design of the plate to facilitate correct positioning of the plate and bone fragments (Fig. 3). To re-create the preoperative simulation during surgery, custom-made guides with guided drill holes and osteotomy cutting slots were designed using commercially available software (3-Matic). These virtual guides were then generated by 3-dimensional printing technology as medical grade polyamide models and sterilized prior to surgery. The plates were 3-dimensionally printed in titanium grade 23 (Ti6Al4V ELI; (LayerWise N.V., Leuven, Belgium) by a selective laser melting installation and were
5 J Hand Surg Am.
1.e3
matrix, 512 512 pixels; pixel size, 0.269 mm; Siemens Healthcare, Erlangen, Germany) from the elbow to the carpometacarpal joints. The CT data were saved in a standard format (DICOM [Digital Imaging and Communications in Medicine]). The formatted files were then imported into the computer simulation software (Mimics and SurgiCase; Materialise N.V., Leuven, Belgium) to segment the CT images and generate 3-dimensional virtual models of both forearm bones of the malunited and mirrored contralateral forearm. These virtual models allow multiplanar visualization and processing with 3-dimensional annotation and measurement. The degree of deformity was evaluated with 3-Matic software (Materialise N.V.) by using the angles between the principal axes of inertia (PAI) of the proximal and distal radius and ulna, adapting techniques previously used and validated by Victor and colleagues14 (Fig. 3). The rotational and translational displacements were calculated as the difference between the affected bone model and the mirror model of the normal bone superimposed proximally to distally. Based on these 3-dimensional models, virtual osteotomies were simulated to align the malunited forearm with the contralateral side, thus producing a precise correction of angular and rotational malalignment and restoring ulnar variance (Video A; available on the Journal’s Web site at www.jhandsurg.org).
15
51 Radius opening wedge Ulna closing wedge M 4
6.6
7.2
Conservative
Midthird
7
41 Radius oblique opening wedge Ulna opening wedge M 3
13.1
14.2
Conservative
Midthird
14
29 Radius closing wedge Ulna oblique F 2
12.8
14.1
Conservative
Midthird
16
51 107 Radius opening wedge Ulna opening wedge Midthird Conservative 13.4 4.4 F 1
Age at Time Fracture (y) Sex Case
TABLE 1.
Patient Characteristics
Age at Time Osteotomy (y)
Initial Treatment
Level of Malunion
Osteotomy
Interval To Osteotomy (mo)
Follow-Up (mo)
FOREARM CORRECTION BASED ON COMPUTER PLANNING
r
Vol. -, - 2017
1.e4
FOREARM CORRECTION BASED ON COMPUTER PLANNING
FIGURE 1: Preoperative A and postoperative B radiographs of a patient with a malunited fracture of both forearm bones.
and guides and V1,500 (U.S. $1,600) for the 2 implants.
FIGURE 2: Patient-specific constructs.
osteotomy
guides
and
Surgical technique Based on the preoperative 3-dimensional virtual planning, all patients underwent an opening, closing, or oblique osteotomy of the radius and ulna to obtain the best possible correction (Table 1). The patient was placed in the supine position with the affected arm on an arm table with a tourniquet applied. In all cases, the ulna was corrected first with a medial approach between the flexor carpi ulnaris and the extensor carpi ulnaris. The polyamide guide was precisely positioned on the malunited ulna using the surface anatomy of the bone (Video A; available on the Journal’s Web site at www.jhandsurg.org). A polyamide 3-dimensional printed copy of the ulna with attached guide and fluoroscopy are used to confirm the correct position of the guide (Fig. 4). The first drill guide was fixed to the ulna with 1.3-mm K-wires and the ulna was drilled through the guiding drill holes. Then the osteotomy guide was placed in position with the use of the same K-wires and the osteotomy cuts were made with an oscillating saw (TPS system; Stryker Orthopaedics, Mahwah, NJ). The guide and K-wires are removed and the patient-specific plate was positioned by inserting the smooth pegs in the predrilled holes, providing an initial reduction and fixation.
plate
designed according to dynamic compression plate principles. Total costs per patient currently amount to approximately V2,500 (U.S. $2,700) for planning J Hand Surg Am.
r
Vol. -, - 2017
FOREARM CORRECTION BASED ON COMPUTER PLANNING
1.e5
FIGURE 3: Study design. Preoperative CT scans are loaded in Mimics A and are applied to perform the 3-dimensioonal planning B for the corrective osteotomy with patient-specific guides and plates. After surgery, a postoperative x-ray is performed C, and the proximal parts of the bone models of the preoperative A, the 3-dimensional planning B, and the postoperative C status are superimposed for the radiographic analysis in the plane of the lateral and AP views of the postoperative radiograph in 3-Matic.
Subsequently, the plate was further fixed with 4 titanium nonlocking, fully threaded cortical screws (2.7 mm diameter in 2 patients, 3.5 mm diameter in 3 patients), providing fixation in 6 cortices both proximally and distally of the osteotomy (Synthes, West Chester, PA). The smooth pegs of the patientspecific plates were 0.5-mm undersized compared with the drill holes in the bone, allowing additional compression by the screws. The correction was verified under fluoroscopy. For the radius, a volar Henry approach was used and a similar procedure was performed. Additional soft tissue release was not required for correction of the deformity.
home exercise program under parent supervision. Full load-bearing and contact sports was allowed once osseous union was well established. Because of the age of the patients, plate removal was planned routinely in all patients. Radiographic evaluation Radiographic evaluation of the angular correction was performed using the preoperative CT-based 3-dimensional models, the planned corrected 3dimensional models, and the postoperative orthogonal radiographs of the osteotomized forearm (Fig. 1B). After surgery, CT scans were not performed to avoid unnecessary radiation of these skeletally immature patients. To allow radiological comparison of the 3 phases (preoperative, planning, postoperative), the 3-dimensional models of the
Postoperative management The forearm was protected with an above-elbow orthosis for 2 weeks, followed by rehabilitation in a J Hand Surg Am.
r
Vol. -, - 2017
1.e6
FOREARM CORRECTION BASED ON COMPUTER PLANNING
procedure a single time for the preoperative, planned, and postoperative phases. In the Results section, the planned angulations (based on the contralateral side) are reported as positive values. The pre- and postoperative angulations are reported positive when the distal bone was deviating in the same direction as the planned angulation. Overcorrection errors are reported as positive and undercorrection errors are reported as negative. In addition, postoperative ulnar variance was assessed by comparing the ulnar variance of the affected side with the contralateral side on the postoperative radiographs. Clinical evaluation An experienced hand therapist (V.B.) performed independent clinical assessments before surgery, at 1 year, and at a minimum of 2 years follow-up of range of motion, grip strength, and pain. Mean follow-up was 42 months (range, 29e51 months). The ranges of wrist flexion, wrist extension, forearm pronation, and forearm supination were measured in degrees with a goniometer. Grip strength was measured in Newtons (N) with a JAMAR isometric handgrip dynamometer (Sammons Preston Rolyan, Bolingbrook, IL). The JAMAR handle was set to position 2 for all subjects. Grip strength was evaluated alternately 3 times for the right hand and 3 times for the left hand with the elbow flexed to 90 and the forearm and wrist in neutral position. The average score was used and a 15% correction for limb dominance was performed for right-handed individuals. Pain scores were measured between 0 (no pain) and 10 (maximum pain) with a visual analog scale.
FIGURE 4: Confirmation of the correct position of the guide by making a fluoroscopy B of the patient’s ulna, together with a polyamide 3-dimensional printed copy of the ulna, attached to the guide A.
preoperative malunion and planned correction were evaluated in the same views as the postoperative radiographs. The DICOM files of the postoperative radiographs and the 3-dimensional models of the preoperative and corrected radius and ulna were imported to the Mimics and 3-Matic software. The 3-dimensional models were set on transparent mode and were individually positioned such that the proximal contours of the 3-dimensional bone models fitted the contours of the radius and ulna on the radiograph (Fig. 3). This alignment procedure was performed once by 3 independent observers (all 3 were orthopedic surgeons), blinded to the measurements of the other evaluators. After aligning the 3-dimensional bone models on the radiographs, the PAI of the proximal and distal radius and ulna were projected on the plane of the radiograph, allowing a quantification of the preoperative, planned, and postoperative angulation of the ulna and radius in the anteroposterior (AP) and lateral planes (Fig. 3). The angulations were then quantified as the angles between the projected PAI of the proximal and the distal part of the radius and ulna. The 3 observers performed this measurement J Hand Surg Am.
Statistical methods Owing to the small sample size, nonparametric statistical methods were employed and the differences between the preoperative and the postoperative range of motion, grip strength, and visual analog scale pain scores were determined using the related samples Wilcoxon signed-rank test. The intraclass correlation coefficient was used to estimate the interobserver reliability of the radiological measurements among 3 independent observers. A P value of .05 or less was considered statistically significant. Ninety-five percent confidence intervals are indicated in the graphical presentation of the clinical and radiographic measurements. Results were expressed as mean (range). Bar graphs were used to indicate means with 95% confidence intervals. r
Vol. -, - 2017
1.e7
FOREARM CORRECTION BASED ON COMPUTER PLANNING
Ulna
Radius
Average Angulation (°) With 95% CI
30
Preoperative Planned Postoperative Correction error
20
10
0
–10
–20
AP
Lateral
AP
Lateral
FIGURE 5: Radiographic measurements. Average angulation of the preoperative, planned, and postoperative ulna and radius, and error in angular correction, measured in the plane of the AP and lateral radiographs with 95% confidence interval (95% CI) bars. Planned angulation was defined positive for all subjects, overcorrection errors are reported as positive, and undercorrection errors as negative values.
RESULTS Radiographic results For the ulna, the preoperative angulation varied between e19.7 and 23.3 in the lateral view and between e13.4 and 20.8 in the AP view. After correction, these deviating angulations were normalized to a range between 0.9 and 6.6 in the lateral view and between 0.6 and 5.2 in the AP view corresponding to the natural curvature of the contralateral bone (Fig. 5). The difference between the planned and the actual correction of the ulna was on average 1.6 (range, 0.6 e4.0 ) in the lateral view and 2.1 (range, 0.3 e5.2 ) in the AP view (Fig. 5; Table S1; available on the Journal’s Web site at www.jhandsurg.org). For the radius, the preoperative angulation varied between e12.5 and 18.1 in the lateral view and between e14.1 and 19.9 in the AP view (Fig. 5). After correction, these deviating angulations were normalized to a range between 1.7 and 9.7 in the lateral view and between 3.1 and 8.4 in the AP view (Fig. 5). The difference between the planned and the actual correction of the radius was on average 1.0 (range, 0.2 e2.3 ) in the lateral view and 1.8 (range, 0.2 e3.3 ) in the AP view (Fig. 5; Table S1; available on the Journal’s Web site at www.jhandsurg.org).
J Hand Surg Am.
The interobserver reliability of the radiological measurements of the alignment of the preoperative, planned, and postoperative models of the ulna and radius was 0.99 (the intraclass correlation coefficient, calculated for 180 observations [3 observers 5 patients 2 radiographic planes 2 bones 3 phases]). For all patients, the postoperative difference in ulnar variance between the affected and the contralateral side was within 1 mm. Clinical results One patient was found to have insufficient osseous union at the time of ulnar plate removal and had immediate replating with a compression plate, followed by successful union. The other 4 patients healed well without complications. All patients showed restricted forearm rotation before surgery and regained close to normal forearm rotation following the osteotomy without release of the interosseous membrane (Fig. 6). The clinical assessments following the corrective surgery showed important improvements in pain score (P < .05), grip strength (P < .05), flexion (P ¼ .07), forearm supination (P < .05), and forearm pronation (P ¼ .08) (Fig. 6).
r
Vol. -, - 2017
1.e8
FOREARM CORRECTION BASED ON COMPUTER PLANNING
FIGURE 6: Clinical measurements. A Pre- and postoperative average scores for pain, range of motion, and grip strength with 95% confidence interval (95% CI) bars. B Pre- and postoperative average scores for range of motion and grip strength compared with the uninjured contralateral forearm with 95% CI bars. A value of 100% means no difference compared with the contralateral side.
DISCUSSION This study investigated the accuracy of the angular correction and clinical outcome of preoperative CTbased 3-dimensional planning and patient-specific guides and plates for the treatment of pediatric forearm malunions. Earlier clinical and cadaver studies have demonstrated that forearm malunions may cause severe restriction of forearm rotation.15e18 Angular deformities in the middle or distal third of the forearm16 and combined multidirectional angular deformities of the radius and ulna17 have been shown to have a J Hand Surg Am.
higher impact on the range of motion than proximal and unidirectional deformities. Multiplanar correction of forearm malunions with restoration of ulnar variance requires meticulous preoperative planning. Translating these plans into an accurate surgical correction can be challenging, and a failure to appreciate the magnitude of the required correction, resulting in inaccurate surgical osteotomies, may lead to inferior clinical results, particularly in complex deformities of the forearm. Previous studies have used 2-dimensional orthogonal radiographs in the preoperative planning.
r
Vol. -, - 2017
FOREARM CORRECTION BASED ON COMPUTER PLANNING
However, assessment of 3-dimensional deformities with 2-dimensional images has significant limitations.2,6,7,9,10,12,18 Owing to difficulty in estimating axial malunions with 2-dimensional radiographs, it has previously been recommended to assess axial malunions before surgery with magnetic resonance imaging and fluoroscopy coupled with goniometry to plan corrective forearm osteotomies.19,20 Nagy et al19 performed orthogonal radiography and axial magnetic resonance imaging as preoperative planning for 17 forearm osteotomy patients. They noted that residual angulation or axial malunion after forearm osteotomies resulted from incorrect planning or inaccuracy in the surgical correction. Whereas the postoperative mean arc of forearm motion improved from 82 to 124 , the overall improvement in motion for patients with predominantly a pronation deficit was poor, and the mean residual restricted forearm motion was 44 . They concluded that combining correction of angular and axial deformities in the forearm was difficult because rotation around an oblique osteotomy induces a change in angulation. More recently, studies have described the use of 3dimensional technology to improve preoperative planning and develop customized cutting guides for osteotomies of the femur, tibia, humerus, and forearm bones.10,11,13,14,21 Murase et al10 investigated 10 patients with malunited diaphyseal forearm fractures who were managed by corrective osteotomy with a custom-made guide based on computer simulations. The average angle of deformity improved from 16 (range, 5 e33 ) before surgery to 1 (range, 0 e3 ) after corrective osteotomy. The average range of forearm pronation and supination significantly improved from 60 and 19 , respectively, to 82 and 73 . One patient, who had corrective surgery 9 years after initial injury, had persistent restricted forearm supination despite good radiological correction. Miyake and colleagues11 reported on 20 forearm malunion patients using the same technique. The average angle of deformity improved from 21 (range, 12 e35 ) before surgery to 1 (range, 0 e4 ) after surgery with an improvement in the mean arc of forearm motion from 76 to 152 . However, forearm supination was still restricted by 70 in 3 patients who had been younger than 10 years at the time of their initial injury and who had long-standing malunion for 96 months or longer. It was surmised that changes in joint configurations and soft tissue contractures during the long delay prior to correction caused residual restrictions in forearm supination. In the current study, we evaluated 5 carefully selected pediatric patients. All patients were skeletally J Hand Surg Am.
1.e9
immature, showed complex malunions of the radius and ulna, and showed incongruence between the standard plates and the bones. Therefore, these patients did not require only customized cutting guides but also patient-specific plates to enhance precise reconstruction of the radius and ulna. For the preoperative planning, we identified the most favorable pattern of corrective osteotomy by superimposing the 3-dimensional models of the malunited and contralateral forearms. Following the planning stage, the patient-specific guides and plates enabled precise intraoperative reconstruction of the preoperative virtual osteotomy simulation, with no intraoperative issues. Compared with the earlier findings of Murase et al10 and Miyake et al,11 the accuracy obtained in this study is very similar. The clinical results of our 5 patients were very good; none of the patients showed restricted forearm range of motion after follow-up, whereas Miyake and colleagues11 still found restricted forearm supination in 3 young patients with long-standing malunions of 96 months or longer. This study also included 1 patient with a longstanding malunion of 107 months who was 4 years old at the time of the injury and 10 at the time of corrective surgery. In this patient, potential changes in joint configurations and soft tissue contractures did not cause any residual restriction of forearm supination. The other 4 patients, aged between 6 and 16 years at the time of injury, had a corrective osteotomy with a delay between 7 and 16 months. The short duration between the malunion and the corrective osteotomy of these patients, as well as their young age, may have contributed to the good outcome of the intervention. To elaborate on the relation between surgical delay and clinical outcome, continued research on larger study groups, including adult patients with long-standing malunions, is necessary. Whereas Murase et al10 and Miyake et al11 used standard plates, we additionally used patient-specific plates. Several studies have demonstrated that the mechanical properties of selective laser melting 3dimensional printed titanium are comparable with, or better than, the standard Ti6Al4V material,22,23 but its use as patient-specific bone plates has only been described for a limited number of surgical applications: for hallux valgus deformities,24 complex pelvic fractures,25 and craniofacial applications.26 The 3-dimensional printed patient-specific forearm plates in this study permitted a perfect fit on the corrected bones and provided the possibility of adding pegs to the plates. These features facilitated the reduction of the osteotomy and allowed precise r
Vol. -, - 2017
1.e10
FOREARM CORRECTION BASED ON COMPUTER PLANNING
fixation of the bony fragments in the planned position. Intraoperative bending of standard plates is an alternative option,21 but is time consuming and the result depends on the complexity of the anatomy and the surgeon’s skill. Whether the clinical benefits of patient-specific implants outweigh the additional cost remains to be determined. The authors acknowledge several weaknesses of this study. In this preliminary series, we assessed the accuracy of corrective osteotomies using 3dimensional technology without comparing the results with a control group without 3-dimensional technology. During the study period, only 5 patients met the inclusion criteria. Because of the small sample size, nonparametric statistical tests were performed that inevitably decreased the power of the study. As a result, this study may fail to detect statistical differences between the pre- and the postoperative situation. Nevertheless, this preliminary study demonstrated statistically significant improvements for several parameters. The contralateral forearm was used to calculate the magnitude of 3-dimensional correction, although forearm-toforearm variability in the normal population is well recognized.20 Future computerized simulations using mathematical models might be developed to avoid imaging the opposite forearm as well as incorporating the biomechanical effects of the interosseous membrane20 but, at present, this is not available. There was a potential bias toward patients with complex malunions who were motivated to try new technology and participate in this study to assess the potential advantage of the 3-dimensional printed patientspecific guides and plates. Interosseous membrane release was not performed; thus, outcomes were entirely dependent on correction of the osseous deformities. To avoid unnecessary radiation, postoperative CT scans were not performed. Instead, orthogonal radiographs were performed that allowed postoperative quantification of the angular correction but not of the torsional correction. Although the results of this small series are encouraging, continued research on larger study groups is required to further evaluate the potential of this technology. We believe this technique represents an exciting and promising new treatment option in the management of complex forearm malunions.
Lucas H.B. Walschot, MD, PhD, who contributed to the manuscript revision. Funding was received from the More Foundation. REFERENCES 1. Hughston JC. Fracture of the forearm in children. J Bone Joint Surg Am. 1962;44(8):1678e1693. 2. van Geenen RC, Besselaar PP. Outcome after corrective osteotomy for malunited fractures of the forearm sustained in childhood. J Bone Joint Surg Br. 2007;89(2):236e239. 3. Fuller DJ, McCullough CJ. Malunited fractures of the forearm in children. J Bone Joint Surg Br. 1982;64(3):364e367. 4. Schmittenbecher PP. State-of-the-art treatment of forearm shaft fractures. Injury. 2005;36(Suppl 1):A25eA34. 5. Schemitsch EH, Richards RR. The effect of malunion on functional outcome after plate fixation of fractures of both bones of the forearm in adults. J Bone Joint Surg Am. 1992;74(7):1068e1078. 6. Jayakumar P, Jupiter JB. Reconstruction of malunited diaphyseal fractures of the forearm. Hand (N Y). 2014;9(3):265e273. 7. Price CT, Knapp DR. Osteotomy for malunited forearm shaft fractures in children. J Pediatr Orthop. 2006;26(2):193e196. 8. Trousdale RT, Linscheid RL. Operative treatment of malunited fractures of the forearm. J Bone Joint Surg Am. 1995;77(6):894e902. 9. Bilic R, Zdravkovic V, Boljevic Z. Osteotomy for deformity of the radius. Computer-assisted three-dimensional modelling. J Bone Joint Surg Br. 1994;76(1):150e154. 10. Murase T, Oka K, Moritomo H, Goto A, Yoshikawa H, Sugamoto K. Three-dimensional corrective osteotomy of malunited fractures of the upper extremity with use of a computer simulation system. J Bone Joint Surg Am. 2008;90(11):2375e2389. 11. Miyake J, Murase T, Oka K, Moritomo H, Sugamoto K, Yoshikawa H. Computer-assisted corrective osteotomy for malunited diaphyseal forearm fractures. J Bone Joint Surg Am. 2012;94(20): e150. 12. Jupiter JB, Ruder J, Roth DA. Computer-generated bone models in the planning of osteotomy of multidirectional distal radius malunions. J Hand Surg Am. 1992;17(3):406e415. 13. Oka K, Murase T, Moritomo H, Goto A, Sugamoto K, Yoshikawa H. Corrective osteotomy using customized hydroxyapatite implants prepared by preoperative computer simulation. Int J Med Robot. 2010;6(2):186e193. 14. Victor J, Premanathan A. Virtual 3D planning and patient specific surgical guides for osteotomies around the knee: a feasibility and proof-of-concept study. Bone Joint J. 2013;95-B(11 Suppl A): 153e158. 15. Morrey BF, Askew LJ, An KN, et al. A biomechanical study of normal functional elbow motion. J Bone Joint Surg Am. 1981;63(6): 872e877. 16. Sarmiento A, Ebramzadeh E, Brys D, Tarr R. Angular deformities and forearm function. J Orthop Res. 1992;10(1):121e133. 17. Tarr RR, Garfinkel AI, Sarmiento A. The effects of angular and rotational deformities of both bones of the forearm. An in vitro study. J Bone Joint Surg Am. 1984;66(1):65e70. 18. Dumont CE, Thalmann R, Macy JC. The effect of rotational malunion of the radius and the ulna on supination and pronation. J Bone Joint Surg Br. 2002;84(7):1070e1074. 19. Nagy L, Jankauskas L, Dumont CE. Correction of forearm malunion guided by the preoperative complaint. Clin Orthop Relat Res. 2008;466(6):1419e1428. 20. Dumont CE, Pfirrmann CW, Ziegler D, Nagy L. Assessment of radial and ulnar torsion profiles with cross-sectional magnetic resonance imaging: a study of volunteers. J Bone Joint Surg Am. 2006;88(7): 1582e1588. 21. Kataoka T, Oka K, Miyake J, Omori S, Tanaka H, Murase T. 3Dimensional prebent plate fixation in corrective osteotomy of malunited upper extremity fractures using a real-sized plastic bone model
ACKNOWLEDGMENTS We are grateful to Kristien Vuylsteke (clinical research coordinator at More Foundation), who took care of patient follow-up and data collection and to
J Hand Surg Am.
r
Vol. -, - 2017
FOREARM CORRECTION BASED ON COMPUTER PLANNING
24. Smith KE, Dupont KM, Safranski DL, et al. Use of 3D printed bone plate in novel technique to surgically correct hallux valgus deformities. Tech Orthop. 2016;31(3):181e189. 25. Wang D, Wang Y, Wu S, et al. Customized a Ti6Al4V bone plate for complex pelvic fracture by selective laser melting. Materials. 2017;10(1):35. 26. Jacobs C, Lin A. A new classification of three-dimensional printing technologies: systematic review of three-dimensional printing for patient-specific craniomaxillofacial surgery. Plast Reconstr Surg. 2017;139(5):1211e1220.
prepared by preoperative computer simulation. J Hand Surg Am. 2013;38(5):909e919. 22. Murr LE, Quinones SA, Gaytana SM, et al. Microstructure and mechanical behavior of Ti-6Al-4V produced by rapid-layer manufacturing, for biomedical applications. J Mech Behav Biomed Mater. 2009;2(1):20e32. 23. Rafi HK, Karthik NV, Gong H, et al. Microstructures and mechanical properties of Ti6Al4V parts fabricated by selective laser melting and electron beam melting. J Mater Eng Perform. 2013;22(12): 3872e3883.
J Hand Surg Am.
1.e11
r
Vol. -, - 2017
1.e12
FOREARM CORRECTION BASED ON COMPUTER PLANNING
TABLE S1.
Clinical and Radiographic Measurements per Patient Patient
Clinical Measurements
1
2
3
4
5
Preoperative
80/55
82/75
45/25
84/31
50/50
Postoperative
90/90
82/90
85/90
90/90
88/85
Preoperative
135.0
157.0
70
115
100
Postoperative
180.0
172.0
175
180
173
Preoperative
32.0
18.3
29.3
10
30
Postoperative
36.0
25.0
36
14.7
30.3
Preoperative
3.5
3.4
1.2
3
5
Postoperative
1.0
0.0
0
0
2.6
e13.4
5.7
11.3
20.8
2.3
Forearm rotation: pronation/supination ( )
Arc of rotation ( )
Grip strength (kg)
Pain scores (VAS)
Radiographic Measurements Ulna—AP view ( ) Preoperative Planned
1.9
6.4
6.3
5.2
1.0
Postoperative
0.6
5.2
1.1
2.8
1.4
Error
1.3
1.3
5.2
2.4
0.3
Ulna—lateral view ( ) Preoperative
7.0
e0.1
e19.7
23.2
6.4
Planned
0.3
5.8
5.2
5.5
0.4
Actual
0.9
6.6
4.3
4.0
4.4
Error
0.6
0.8
0.9
1.5
4.0
Preoperative
16.2
e5.5
5.7
e14.1
19.9
Planned
9.5
4.9
8.0
0.4
4.6
Actual
8.4
7.0
4.7
3.1
4.4
Error
1.1
2.1
3.3
2.7
0.2
Preoperative
9.5
11.2
18.1
e12.5
3.0
Planned
4.2
5.4
2.4
10.0
4.3
Actual
4.7
3.1
1.7
9.8
5.6
Error
0.5
2.3
0.7
0.2
1.3
Radius—AP view ( )
Radius—lateral view ( )
VAS, visual analog scale.
J Hand Surg Am.
r
Vol. -, - 2017